The first project area explores metabolic pathways that have been proposed based on in vitro studies to be important in non-replicating (NR)-TB. We have previously provided chemical and genetic validation of the terminal step in NAD biosynthesis, NAD synthase (NadE), as a drug target for both replicating and non-replicating Mycobacterium tuberculosis. This study is being pursued to provide further validation of NadE as a drug target by fragment-based drug development approaches where small molecules identified as binders in in silico docking studies of NadE as well as compounds developed from a previous series of indole inhibitors are tested for inhibition of NadE enzyme activity. In addition, survival of mutants of M. tuberculosis where NadE levels are controlled by a tetracycline-regulatable promoter are being tested for survival under a variety of non-replicating conditions. Other pathways that we have previously explored for importance in NR-TB include biotin biosynthesis, peptidoglycan turnover and iron acquisition and work is continuing to validate these by chemical and genetic means. We have been able to demonstrate that M. tuberculosis peptidoglycan has interpeptide crosslinks which are proportionally different to those of bacteria such as E. coli and that these linkages and their relative amounts are surprisingly similar between different stages of growth and non-replicating persistence. The identity of the transpeptidases that play a role in crosslinking under the different conditions is being explored by analysis of genome sequences of clinical strains with differential beta-lactam susceptibilities, affinity pull-down and generation of mutants. The second major focus area of this project starts from a different perspective and uses compounds that are in clinical development (PA-824 and SQ109) that are known to possess activity against NR-TB and attempts to understand how they work. Building on our elucidation of the mechanism of action of PA-824 against NR-TB, we are analyzing nitric oxide release and the relative amounts of other metabolites produced during the reduction of PA-824 by the Rv3547 nitroreductase (Ddn). In addition we are exploring the native function of Rv3547 in the cell and have determined that it may be a DNA repair enzyme that removes 8-nitro-Guanine from DNA damaged during nitrosative stress. The third major focus of this project involves global approaches to understanding the metabolism in NR-TB. We have developed a chemostat model of M. tuberculosis for growth of the organism under defined oxystatic conditions. We have determined the lower oxygen concentration limits at which measurable growth can still occur and are exploring metabolism by a combination of metabolomic, transcriptional and metabolic inhibitor sensitivity studies. In a second approach we are identifying inhibitors of metabolism by high-throughput screening approaches which would have as ultimate goal the mapping of metabolism under defined environmental growth or NR conditions. To date we have completed 8 high-throughput screens of M. tuberculosis growing on defined carbon sources against a library of 15,000 compounds from the NCGC collection. Understanding the differential susceptibility of M. tuberculosis to inhibitors under these growth conditions will help define the relative role of the various metabolic pathways. Hits that are common to all or some screening conditions or unique to only one condition are now being followed up by chemoinformatic analyses of all available literature on these compounds and chemically similar analogs. Mutant selection and transcriptional profiling will be used as a first approach to identify the targets of attractive hits. A similar approach is being used to follow up on hits from a high-throughput screen of M. bovis BCG persisting under anaerobic conditions done in collaboration with NITD and GNF. The latter screen was designed to identify respiratory inhibitors since we have established that respiration is a bottleneck in survival of such non-replicating mycobacteria and in addition to the above approaches, we are also analyzing susceptibility of respiratory mutant strains to these hits and related analogs under a variety of conditions. Unfortunately 15,000 compounds do not provide the chemical diversity to explore a significant fraction of metabolism. Moreover, no synthetic library would cover the chemical space required to map metabolism. So another aspect of this project has been to explore alternate sources for compounds that might be enriched in molecules with activity against NR-TB. One source that has been especially interesting is samples that have been collected from sphagnum peat bogs. Mycobacteria present in this environment face a very similar set of conditions to those within the diseased lung of a human TB patient. Competitor bacteria from the core of such bogs have been identified and characterized and some produce compounds of remarkable potency against NR-TB. Not surprisingly, a large fraction of the antibiotic producing bacteria are either unknown bacteria or bacterial genuses not previously associated with antibiotic production. The mechanism of action of bog extracts that inhibit M. tuberculosis growth or survival is being analyzed by transcriptional profiling. Microarray analyses have proven to be a rapid method of eliminating those bog extracts containing an inhibitor that targets previously known metabolic targets such as protein synthesis, DNA integrity and iron acquisition.